RU100347U1 - DEVICE FOR SCANNING OBJECTS - Google Patents

Abstract

 1. A device for scanning objects containing a light source, an optical system for equalizing the amplitude of the light emitted by the source over the spectral range, an optical system for decomposing the light flux into spectrum components and a photosensitive sensor in the form of a CCD matrix, characterized in that the CCD matrix includes, at least two layers of photosensitive elements configured to absorb photons of the light flux by the photosensitive elements of the first, relative to their movement, CCD layer Ritsa and / or at least one of its layers during subsequent breakdown photons previous layers and storage layers of the photosensitive elements of electric charges, wherein the matrix is provided with electrodes for the parallel charge transfer its overflow from one layer to another. ! 2. The device according to claim 1, in which the CCD matrix has at least two regions configured to absorb in each of them a portion of the spectrum of the decomposed light flux. ! 3. The device according to claim 1, in which each of the mentioned region of the CCD layer of the matrix is made in the form of at least one row of active pixels from photosensitive elements. ! 4. The device according to claim 1, in which the output elements are installed in the last layer of the CCD matrix, and the pixels of each row in the last layer are connected by successive charge transfer electrodes to transfer the accumulated charge from all layers in each row to the output elements. ! 5. The device according to claim 4, further comprising an analog-to-digital converter connected to the output elements of the CCD matrix, and a storage device connected to it and

Description

The utility model relates to photoelectronic technology, and in particular, to devices for scanning objects that can be used, in particular: for scanning any color information carriers, both for transparency and reflection (for example, film, film, etc.) scans of any material objects (both in light and reflection), including any celestial bodies, as well as cells, particles, molecules, living tissues, for flaw detection of materials and products, for photo, film and video shooting, in t .h. observation video for shooting and recording holographic objects, for observing and recording the degree of illumination of material objects and structures based on them.

In general, the task of scanning is to determine the spectral and brightness characteristics of the light flux corresponding to the characteristics of the scanned object.

The prior art device for scanning objects (see EP 1564985 A2, 08/17/2005) containing a light source, a liquid crystal matrix for decomposing the light flux and forming bands of red, green and blue light (RGB) and a photosensitive sensor in the form of a CCD.

A disadvantage of the known device is that the analysis of the luminous flux, which carries information about the scanned object, is performed only in three regions of the spectrum (RGB), while the remaining regions are not taken into account. In addition, the CCD matrix used in this device does not provide “capture” of all photons of light that hit its surface.

The objective of the claimed utility model is to create a device for scanning an object that allows you to get the highest quality digital image.

The technical result of the claimed utility model consists in the possibility of determining the intensity of the light flux passing through the scanned object (or reflected from it) in any part of the spectrum in the field of visible radiation and adjacent areas (infrared or ultraviolet), as well as in improving the accuracy of these data .

The specified technical result is achieved due to the fact that the device for scanning objects contains a light source, an optical system for aligning the amplitude of the light emitted by the source over the spectral range, an optical system for decomposing the light flux into spectral components and a photosensitive sensor in the form of a CCD matrix including at least two layers of photosensitive elements configured to absorb photons of the light flux by photosensitive elements of the first tionary their movement CCD layer and / or at least one of its successive layers in the breakdown by photons previous layers and storage layers of the photosensitive elements of electric charges, wherein the matrix is provided with electrodes for the parallel charge transfer its overflow from one layer to another.

In addition, the specified technical result is achieved due to the fact that:

- the CCD matrix has at least two regions configured to absorb in each of them a portion of the spectrum of the decomposed light flux,

- each region of the CCD layer of the matrix is made in the form of at least one row of active pixels from photosensitive elements,

- in the last layer of the CCD matrix, output elements are installed, and the pixels of each row in the last layer are connected by electrodes of sequential charge transfer, to transfer the accumulated charge from all layers in each row to the output elements,

- the device further comprises an analog-to-digital converter connected to the output elements of the CCD matrix, and an information storage device connected to it,

- the device further comprises a slit mask for focusing the luminous flux, made with the possibility of placing in its center a scan object,

- the light source is made in the form of a LED matrix with feedback,

- The TDI matrix is used as the CCD.

The preferred embodiment of the claimed device for scanning is shown in figure 1

The scanning device comprises a light source 1 (for example, a feedback LED array), an optical system 2 for normalizing the light flux (for equalizing the light amplitude over the spectral range), a slit mask 3, an optical system 4 (channel, path, prism lens system, gaps and gratings) for decomposing the light flux into spectrum components, a CCD matrix 5, an ADC (analog-to-digital converter) 6 connected to the output elements of the matrix and an information storage device 7 connected to the ADC 6.

In general, a CCD matrix can be an analog integrated circuit consisting of photosensitive elements (photodiodes) and using CCD technology - charge-coupled devices. The matrix can be made of polysilicon (or other material), separated from the substrate (for example, silicon), in which when voltage is applied through polysilicon gates, the electric potentials near the electrodes change. When voltage is applied to the electrodes, a potential well is created in which electrons can accumulate as a result of exposure to light during exposure. The more intense the light flux during exposure, the more electrons accumulate in the potential well, respectively, the higher the total charge of this pixel. Thus, when scanning, the luminous flux, which carries information about the scanned object, is directed to the photosensitive surface of the CCD elements, the task of which is to convert the photon energy into an electric charge. In general, this happens as follows. For a photon that has fallen on a CCD element, there are three options for the development of events - it will either “ricochet” from the surface, or will be absorbed in the bulk of the semiconductor (matrix material), or “will pierce through” its “working zone”. Photons that were absorbed by the matrix form an electron-hole pair if interaction with the atom of the semiconductor crystal lattice occurs, or only an electron (or hole) if the interaction was with atoms of donor or acceptor impurities. However, those photons that “ricocheted” or “pierced” the matrix through and through cannot be taken into account when determining the intensity of the light flux. Obviously, it is necessary to create a matrix in which the losses from the “rebound” and “lumbar sweep” would be minimized.

This problem is solved by using a multilayer matrix 5 used in the claimed device. Its design is shown in figure 2. Unlike known analogues, the matrix contains not one, but at least two photosensitive layers 8 of active pixels 9. Moreover, each of the layers has at least two regions 10, each of which absorbs the decomposed luminous flux 11 in different parts of the light spectrum. Preferably, these areas are made in the form of one or more lines. For electron transfer from layer to layer, the matrix is equipped with electrodes 12 for parallel charge transfer. At the same time, on the last layer of the matrix, the pixels of each row (region) of the matrix are connected by electrodes 13 for sequential charge transfer, in order to "flow" the accumulated charge on all layers in each region to the output elements 14 of the matrix.

The preferred architecture of the multilayer pixel 9 in the inventive matrix is shown in figure 3.

The pixel 9 is a set of photosensitive semiconductor elements 15 placed in a multilayer substrate 16. A lens 17 is installed in front of the first layer (if reflected light is used) and a transparent electrode 18 is separated from the first layer by an insulator 19. Each matrix layer has a generation zone charge carriers 20 and the zone of the potential well 21 In this case, the layers are separated from each other using transparent or translucent layers 22.

It is most preferable to use a TDI matrix in the inventive device, since it will allow to best record the entire spectrum of incident radiation energies. In this case, each layer will better accumulate and transfer the energy of interaction between photons and electrons, and the total charge of this interaction will be estimated and fixed.

To describe the expansion of the spectral composition of the recorded light radiation, we turn to the description of color spaces, more precisely, their models.

one). RGB space (Fig. 4): Red, Green, Blue - red, green, blue. The color is divided into 3 characteristics expressing the content of the primary colors. The model is additive, since these components are summed. This color space is used when displayed on the monitor screen. This means that the model is hardware dependent, on different monitors the same colors will look different. RGB color is used with different accuracy: 8-bit RGB gives 256 colors, 16-bit 65536 (scheme 5-6-5), 24-bit 16777216 (8-8-8). The brackets indicate the bits per channel.

2) CMYK space (Fig. 5): Cyan, Magenta, Yellow, Key - cyan, magenta, yellow, key (black). This format is used in printers. Saves ink. Unfortunately, you cannot create inks similar to RGB for printing. The thing is that these colors work only “in the light”, i.e. through a film-filter or phosphor monitor. Colors are as if cut out by appropriate filters from a continuous spectrum. In print, everything happens exactly the opposite, i.e. paper absorbs the entire spectrum except for the color in which it is painted. It is not possible to create paints that are absolutely exactly "opposite" (complementary) to RGB colors, so you have to enter the fourth additional paint - black. Its task is to enhance the absorption of light in dark areas, to make them as black as possible, i.e. increase the tone range of printing. The four-channel CMYK is heavier than RGB and processed more slowly, taking up more memory.

3). HLS space (Fig.6): Hue, Lightness, Space - hue, brightness, saturation. A fairly common format, convenient for applying various effects. Unlike the previous two cubic spectra of RGB and CMYK, HLS is conical. The HSB (Hue, Space, Brightness) and HSV (Hue, Space, Value) models, which are also conical, are very similar to it. These patterns are closest to human color perception. In addition, it is most convenient for optical and photometric calculations: the hue corresponds to the wavelength, brightness to the amount of light, saturation to intensity. So this model will be convenient when working with light sources and materials.

four). CIE XYZ space (Fig. 7): Normal color scheme - flat color rendering model. The red color components are elongated along the X axis of the coordinate plane (horizontal), and the green color components are elongated along the Y axis (vertically). With this method of presentation, each color corresponds to a certain point on the coordinate plane. The spectral purity of colors decreases as you move along the coordinate plane to the left. But this model does not take into account brightness. This model is hardware independent, it supports much more colors than modern devices (scanners, monitors, printers can distinguish). CIE XYZ is built on the basis of the visual capabilities of the so-called “Standard Observer”, that is, a hypothetical viewer whose capabilities have been carefully studied and recorded during conducted by the CIE committee of lengthy studies of human vision. The CIE committee conducted many experiments with a huge number of people, asking them to compare different colors, and then using the combined data of these experiments built the so-called color matching functions and the universal color space in which the range was presented visible colors, characteristic of the average person. The color matching functions are the values of each primary light component that must be present so that a person with average vision can perceive all the colors of the visible spectrum

5). CIE Lab Space (FIG. 8): Enhanced XYZ Model. The ultimate goal of the CIE was to develop a repeatable system of color rendering standards for manufacturers of paints, inks, pigments and other dyes. The most important function of these standards is to provide a universal scheme within which color matching can be established. The basis of this scheme is the Standard Observer and the XYZ color space, however, the unbalanced nature of the XYZ space, due to the fact that a person distinguishes between the shades of green and yellow much better than between the shades of red and purple, made these standards difficult to implement clearly. As a result, CIE developed more uniform color scales - CIE Lab and CIE Luv. Of these two models, the CIE Lab model is more widely used. The well-balanced Lab color space structure is based on the theory that color cannot be both green and red or yellow and blue. Therefore, the same values can be used to describe the red-green and yellow-blue attributes. When color is represented in the CIE Lab space, the L value indicates luminosity, a is the value of the red-green component, and b is the value of the yellow-blue component.

There are also other color models (such as CCY, Luv, Mansell and Ostwald models) but they are used much less frequently.

As can be seen from the presented models, the most limited and accordingly dependent space model is the RGB space model, and since Since all existing matrices register light emission precisely according to this model, the luminance characteristic L of these matrices is rigidly tied to color characteristics and any change in the luminance range immediately leads to a change in color and vice versa.

In our case, because the fierce characteristic is measured separately from the color, and the color range model most closely matches the CIE Lab model representation, the range of measured colors and their shades is limited only by the number of steps for recording the color spectrum for a given material and matrix execution. Thus, a fundamentally different level of accuracy is achieved when processing color information and, accordingly, the range of measured and recorded color shades is limited only by the mathematical apparatus of the current representation of the CIE Lab space model.

The claimed device operates as follows.

For scanning in the region of visible radiation, white light is used, and in the case of scanning in the adjacent areas, infrared or ultraviolet light is used. The light is emitted by the light source 1 and normalized using an optical system 2, at the output of which the light is a stream of white light or close to it, whose amplitude (brightness) is aligned over the entire frequency (spectral) range. Further, the luminous flux passes through the slit mask 3, where it is converted into a narrow light beam. A scanning object (for example, a film or a plate) is placed in the center of the slit mask, the plane of which moves perpendicular to the plane of the narrow light flux. Passing through this scanning object to the lumen or reflecting a normalized light flux from it, it changes its spectral composition and brightness character, namely, it acquires a brightness characteristic (amplitude) for each wavelength included in the visible (400-700 nm) or close to it light spectrum, in accordance with the scanned area of the object and depending on the color mask applied to it. The luminous flux thus formed gets into the optical system 4 and, passing through this optical channel in the part of the visible spectrum (400-700 nm) or close to the visible one, differentiates (decomposes, scatters in space due to different refraction angles for different wavelengths) by countless many light waves (spectral components). In this form, the luminous flux enters the photosensitive surface of the CCD of matrix 5. In this case, the decomposed luminous flux is distributed over the surface of the matrix 5 in such a way that at least two regions are formed, in each of which one of the parts of the expanded spectrum is absorbed, i.e. the light flux with a certain wavelength is absorbed. As mentioned earlier, when photons hit the surface of the matrix, most of them interact with photosensitive elements, forming electrons and accumulating them in the zone of potential well 21. Moreover, some photons can pass through one or several layers of the matrix 8, in which case photons will be absorbed by one of the subsequent layers. Thus, a charge will also appear on the intermediate layers of the matrix.

Then, by means of electrodes 12 for parallel charge transfer, the formed electrons from all layers “flow” to the last (relative to the movement of the light flux) layer, in which using electrodes 13 of sequential transfer in each row, the charge from all layers moves with the output elements 14 of the matrix. In this case, the total charge formed in each region of the matrix on all layers determines the intensity (luminance characteristic) of the light radiation carrying information about the scanned object in each part of the spectrum (for each wavelength range).

Figure 9 presents a graph of the characteristic curve (HK) sensitometry. Sensitivity (or density in the case of comparison with film or photo film) is plotted on the graphs along the Y-axis (abscissa), and the exposure or time at which occurs on the X-axis (ordinates), the process of charge accumulation under the influence of incident photons is recorded. The exposure value is measured in lux per second and has a logarithmic form for compactness of convenience of assessment and perception of final values. The values of the so-called a neutral gray standard wedge, according to which the density of blackening (underexposure) in the lower part of Kh.K. is currently measured or whitening (overexposure) in the upper part of HK, allow us to best describe the process of interaction of the number of incident light photons and their energies (and, accordingly, the spectral composition of the recorded light flux) on the registration process itself. The figure shows that the expansion of the dynamic range is achieved in the lower part of H.K. - due to the shift in the area of blackening (underexposure) Kh.K. to the lower part and registration of almost all photons (this happens due to the construction of the very first and most sensitive layer of the matrix described above), by almost an order of magnitude, and in the upper part of Kh. by adjusting the border of whitening (overexposure) i.e. smooth bias of the registration region or reaction to the incident photon flux and the limitations of this flux in the upper part of Kh.K. (this happens roughly - due to the selection of the photosensitive layer material, the amount of impurities in the material, i.e., changes in the number of charge carriers, as well as changes in the size and thickness of both the layers themselves and the light-recording regions of the matrix, and more precise adjustment and adjustment is carried out by shifting the magnitude of the potential applied to the electrode layer and, accordingly, shifting the reaction zone of the potential well towards larger or smaller values). In the “white” region, all the pixels of a conventional matrix will be illuminated and it will be problematic to determine, for example, 10,0001 photons out of 100,000. In the black region, on the contrary, to determine single photons, pixels with high sensitivity are required and it is necessary to distinguish, for example, 3 from 4 fallen photons. Using a multilayer matrix, it is possible to vary the information retrieval from each layer and use, for example, the first layer as a layer that transmits almost all photons and is a layer of error correction. Or vice versa, use it as a highly sensitive layer. You can also change the sensitivity of the layers relative to each other, creating, for example, a logarithmic distribution of sensitivity between the layers. Thus, it is possible to expand the brightness range in which the claimed matrix operates. Existing matrices work, as a rule, only in the L or Lmax zone.

Thus, through the use of an optical system for preliminary decomposition of the light flux into spectral components, the claimed device allows you to determine the intensity of the light flux carrying information about the scanned object in any part of the spectrum. Moreover, the use of a matrix having many layers makes it possible to determine this intensity most accurately due to the maximum absorption of light photons by the matrix layers.

The claimed device allows you to get a better digital image due to:

1) increase the distinguishability of the gradations of gray (up to 10 times) at the same power or illumination of the object, or get the same quality of distinguishability with illumination of lower illumination than by known methods. The dynamic luminance coefficient of distinguishability D (the ratio of the lightest to the darkest) using the claimed method is D = 4, and the distinguishability is 10000, using similar methods - D = 3, the distinguishability is 1000.

2) A significant increase in the distinguishability of the number of color shades (up to 16,000,000 times) compared with existing technologies. Using existing methods, the distinguishability of spectrum 2 to 24 degrees (linear system). Using the proposed method, the distinguishability of spectrum 2 to 48 degrees (logarithmic system).

The described construction can be implemented, in particular, in the following devices: cinema scanners, movie cameras, television cameras, video cameras, cameras, digital surveillance cameras, motion cameras, night vision cameras, etc., digital binoculars, telescopes, telescopes, electron microscopes, flaw detectors, copiers, parking sensors, medical probes, devices for “artificial vision”, manual input devices, geodetic recording devices, meteorological recording devices, devices for astronomical and aerial photography, stereo and 3D scanners, workflow scanners, feedback devices, in object registration systems, devices for monitoring and recording illumination of objects of material structure.

It should be noted that the design of the scanning device disclosed in the description is the preferred embodiment of the claimed utility model, but not limiting the set of essential features provided in the application. Moreover, all of the specified set is aimed at achieving the aforementioned technical result.

Claims (8)

1. A device for scanning objects containing a light source, an optical system for equalizing the amplitude of the light emitted by the source over the spectral range, an optical system for decomposing the light flux into spectrum components and a photosensitive sensor in the form of a CCD matrix, characterized in that the CCD matrix includes, at least two layers of photosensitive elements configured to absorb photons of the light flux by the photosensitive elements of the first, relative to their movement, CCD layer Ritsa and / or at least one of its layers during subsequent breakdown photons previous layers and storage layers of the photosensitive elements of electric charges, wherein the matrix is provided with electrodes for the parallel charge transfer its overflow from one layer to another. 2. The device according to claim 1, in which the CCD matrix has at least two regions configured to absorb in each of them a portion of the spectrum of the decomposed light flux. 3. The device according to claim 1, in which each of the mentioned region of the CCD layer of the matrix is made in the form of at least one row of active pixels from photosensitive elements. 4. The device according to claim 1, in which the output elements are installed in the last layer of the CCD matrix, and the pixels of each row in the last layer are connected by successive charge transfer electrodes to transfer the accumulated charge from all layers in each row to the output elements. 5. The device according to claim 4, additionally containing an analog-to-digital converter connected to the output elements of the CCD matrix, and an information storage device connected to it. 6. The device according to claim 1, additionally containing a slit mask for focusing the light flux, made with the possibility of placing in its center a scan object. 7. The device according to claim 1, in which the light source is made in the form of a feedback LED matrix. 8. The device according to claim 1, in which the TDI matrix is used as the CCD matrix.